Feasibility of Fuel Cell APUs for Automotive Applications

By Chris Spangler, Eric Sattler, and Heather McKee

AET5110 Fuel Cell Fundamentals

Professor Simon Ng

Due 12/05/05

Report Documentation Page Form ApprovedOMB No. 0704-0188Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.

1. REPORT DATE 07 DEC 2005

2. REPORT TYPE Journal Article

3. DATES COVERED 29-08-2004 to 30-11-2005

4. TITLE AND SUBTITLE Feasibility of Fuel Cell APUs for Automotive Applications

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) Chris Spangler; Eric Sattler; Heather McKee

5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) U.S. Army TARDEC,6501 East Eleven Mile Rd,Warren,Mi,48397-5000

8. PERFORMING ORGANIZATIONREPORT NUMBER #15383

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) U.S. Army TARDEC, 6501 East Eleven Mile Rd, Warren, Mi, 48397-5000

10. SPONSOR/MONITOR’S ACRONYM(S) TARDEC

11. SPONSOR/MONITOR’S REPORT NUMBER(S) #15383

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

14. ABSTRACT Fuel cells are currently being evaluated for many applications, including portable power transportation,and large stationary systems. The US military is looking at fuel cells to help reduce its use of fuel in thebattlefield, and to more adequately address the vehicle electrical power demands not being fulfilled bybatteries and on-board generators. Hydrogen will not be used in the battlefield for a long time, if ever, andso the primary concern for introduction of military vehicle fuel cells is on-board fuel processing. Because ofthe military’s decision to implement the Single Fuel Forward policy, JP-8 (and JP-5) is the primary fuelused in the battlefield. When acquired outside of the US where fuel quality is not necessarily regulated, thiscan lead to sulfur levels of up to 3000 ppm, which are several orders of magnitude above the tolerance ofcurrent fuel cell and reformer systems. The Army has several fuel cell programs for vehicle applicationscurrently underway, ranging from successful processing of JP-8 to vehicle demonstration programs, withthe hope that those two areas will eventually converge during the technology evolution process.

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT

Public Release

18. NUMBEROF PAGES

67

19a. NAME OFRESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

2

ABSTRACT

Fuel cells are currently being evaluated for many applications, including portable power,

transportation, and large stationary systems. The US military is looking at fuel cells to

help reduce its use of fuel in the battlefield, and to more adequately address the vehicle

electrical power demands not being fulfilled by batteries and on-board generators.

Hydrogen will not be used in the battlefield for a long time, if ever, and so the primary

concern for introduction of military vehicle fuel cells is on-board fuel processing.

Because of the military's decision to implement the Single Fuel Forward policy, JP-8

(and JP-5) is the primary fuel used in the battlefield. When acquired outside of the US,

where fuel quality is not necessarily regulated, this can lead to sulfur levels of up to 3000

ppm, which are several orders of magnitude above the tolerance of current fuel cell and

reformer systems. The Army has several fuel cell programs for vehicle applications

currently underway, ranging from successful processing of JP-8 to vehicle demonstration

programs, with the hope that those two areas will eventually converge during the

technology evolution process.

3

TABLE OF CONTENTS

Page 4

Page 5

Page 5

Page 6

Page 7

Page 11

Page 11

Page 14

Page 23

Page 25

Page 29

Page 31

Page 35

Page 37

Page 39

LIST OF FIGURES AND TABLES

BENEFITS AND DRAWBACKS OF AUXILIARY POWER UNITS

MILITARY VEHICLE MAIN ENGINE ISSUES

SILENT WATCH AND COMMERCIAL VEHICLE IDLING REDUCTION

BATTERY ISSUES

SINGLE FUEL FORWARD

CLEAN FuELS INITIATIVE

REFORMING FUEL

MILITARY JP-8 REFORMER

HYDROGENICS MULTI-SERVICE REGENERATIVE ELECTROLYZER FuEL

CELL

DELPHI SOFC APU wi REFORMER

FREIGHTLINER TRACTOR WITH BALLARD PEM APU AND METHANOL

REFORMER

SUNLINE TRACTOR WITH HYDROGEN-FuELLED HYDROGENICS PEM

APU AND VEHICLE ELECTRIFICATION

CONCLUSIONS

REFERENCES

4

LIST OF FIGURES AND TABLES

FIGURES LIST

Page 8 Page 8 Page 8 Page 9 Page 10 Page 25 Page 26 Page 26 Page 27 Page 27 Page 29

Page 30 Page 30

Page 31

Page 32

Page 33 Page 33

Page 33

Page 34 Page 35

Page 35

TABLES LIST

Page 10 Page 24 Page 24 Page 24

Page 26 Page 27

Figure 1 - 6TL Battery Figure 2 - 6TLFP Battery Figure 3 - 6TMF Battery Figure 4 - VRLA Battery Figure 5 - Battery graveyard from current use. Figure 6 - MREF Phase II: 3kW PEM Figure 7 - MREF Phase III: 7kW PEM Figure 8 - MREF Phase III: 7kW Stack Figure 9 - MREF Phase III: Electro1yzer Figure 10 - Metal Hydride Tanks Figure 11 - Delphi SOFC APU with integrated fuel processor. The

picture on the right illustrates the volume compared to a typical battery.

Figure 12 - Schematic layout of the current BFV powertrain. Figure 13 - Schematic layout of the current BFV battery and generator

configuration. Figure 14 - Bradley electrical power requirements increasing over time

(provided by the US Army TACOM Bradley PM office). Figure 15 - A graphical representation of the logistics burden of the·

military. Figure 16 - Freightliner Class 8 tractor with the Ballard integrated APU Figure 17 - APU installed on the side framerails ofthe tractor. The fuel

tank is mounted on the opposite side of the truck. Figure 18 - The Ballard module assembly. The stack is on the top, the air

system is on the bottom right, and the fuel processor is on the left.

Figure 19 - Graphical representation of the fuel processor. Figure 20 - Data generated by the DOE NREL ADVISOR model,

comparing the idling of a main diesel engine to this APU configuration.

Figure 21 - The SunLine tractor after its cross-country trek from southern California to Washington, DC, in 2005.

Table 1 - 6TMF/VRLA Comparison of Technologies Table 2 - Comparison of Usage of JP-8 and Cost per 12.5 kWh Table 3 - Comparison Usage of Cost to Charge Battery Table 4 - Scenario: 100 Charge Cycles (JP-8 reformer and fuel cell APU

$28,500 estimate 2010) Table 5 - MREF Stack Specifications Table 6 - MREF Electrolyzer Specifications

5

BENEFITS AND DRAWBACKS OF AUXILIARY POWER UNITS

Fuel cell auxiliary power units (APU) offer many benefits over the performance of a

vehicle's main engine, whether it is for engine idling reduction in the heavy-duty

commercial vehicle world or Silent Watch of military vehicles. APUs have significantly

improved fuel economy; reduced tailpipe emissions; and reduced thermal, visible smoke

and noise signature. Even in the situation in which one requires a fuel reformer, to

process a hydrocarbon fuel such as diesel into hydrogen, the emissions would be less

because the amount of fuel consumed is decreased.

Some of the drawbacks to APUs, whether they are engines or fuel cells, are their added

cost, weight, and maintenance burden. Issues also arise regarding the packaging space

they claim, as well as mechanical and electrical integration. Few, if any, military tactical

vehicles have engine APUs. Combat vehicles such as the Bradley Fighting Vehicle and

the M113 Armored Personnel Carrier also usually lack APUs, but many M1 Abrams

tanks have had small diesel generators retrofitted. Retrofitted APUs are not easily

designed to be protected, as is other integrated hardware on the vehicle. [1] [2]

MILITARY VEHICLE MAIN ENGINE ISSUES

The typical diesel propulsion engine of military vehicles is not well suited for idling over

long periods of time, even though that is one of the most common operating modes in the

battlefield. During extended idle, poor diesel combustion and over-fueling in cold

weather conditions can occur. This can result in diluted oil and engine damage. Engines

are also typically extremely inefficient at idle speeds, on the order of 25% efficiency, or

approximately one to three gallons diesel fuel per hour for trucks. Armored vehicles may

consume up to eight gallons per hour due to greater engine fan power requirements and

larger engines. The Abrams Ml tank, with the propUlsion gas turbine engine, has an

especially poor consumption of over twelve gallons per hour, depending on its idling

speed. [1]

6

SILENT WATCH AND COMMERCIAL VEHICLE IDLING REDUCTION

Silent Watch mode is when a military vehicle and crew are alert and actively monitoring

the battlefield in a scouting or ambush mode. The vehicle batteries provide electrical

power until they reach minimum capacity, and the diesel engine is restarted to recharge

those batteries. If the battery discharge is too low, a risk is run of not being able to restart

the engine at all. Repeated deep discharges will also wear out the batteries. Two other

alternatives for standby electrical power in military vehicles include idling the main

engine continuously and/or using a diesel or gasoline internal combustion engine APU

These options, in addition to battery usage with repeated engine start-up, are considered

poor choices. Since deep discharging dramatically decreases the life of the batteries and

the earsplitting idling engine noise defeats the purpose of Silent Watch, and thus the

motivation for fuel cell APUs, with their better efficiency and fuel processing capability

of the aviation Single Battlefield Fuel. Full-time engine idling is still currently the

dominant means of ensuring electrical power for the communications equipment. [1] [3]

In the commercial trucking and transit bus world, there is increasing emphasis in

eliminating unnecessary idling, such as when long-haul semi-trucks idle overnight at a

truck rest stop. Although this is in a commercial application, the needs are similar to that

of Silent Watch: minimize exhaust emissions, minimize the fuel wasted when idling, and

minimize noise/vibration of the vehicle, all while providing power for hotel loads, such

as heat and air conditioning. The most important requirements are that a technology(s)

has a payback period of less than two years, and that the hardware has a seamless

integration, to minimize changing their habits (other than unnecessary idling). DOE's

Argonne National Laboratory estimates that around 500,000 long-haul trucks idle

between three and sixteen hours per day, with an associated annual fuel cost of around

$3000-4000 per vehicle. There is also a strong cultural belief in the industry that idling

overnight is necessary for ease of cold start-up. Inconsistent regulations; inconsistent

enforcement of the regulations; variability of drive cycles, drivers and applications; and

lack of education of the anti-idling options available to drivers and fleets have, until

recently, prevented progress from being made in this area. In 2004, several government

agencies, including the DOE, EPA, DOT, and DOD, joined forces to work with industry

7

to address these issues, moving idling reduction technology into the policy mainstream,

and are currently finalizing a comprehensive action plan.

Idling reduction technology(s) can be mass-produced at reasonable cost through the

competition and high-volume of the industry. The extremely high mileage accumulation

of typically 100,000 miles per truck per year provides quick durability results. The larger

fleets turnover vehicles every five to ten years, depending on the current market resale

value for used trucks. However, the typical Army truck travels only about 3,000 miles

per year, and the vehicle is typically kept for 20 years or more-this low mileage

accumulation and vehicle tum-over leads to slow or no technology growth. This is a

prime example of the US Army's National Automotive Center's mission of helping

develop dual-use technology for both military and commercial vehicles, and some· of

these programs are outlined below. [1]

BATTERY ISSUES

Batteries are an integral part of today's society. While it is not immediately evident,

batteries are used by people multiple times every day. Examples of this include the latest

portable devices, children's toys, and even most wrist watches. The US Army has

utilized battery technology to power its forces since the early 1950's. There have been

five different battery types used: 6TN, 6TL, 6TLFP, 6TMF, and VRLA.

The first battery to be used was the 6TN battery. This battery was a lead-acid chemistry

with antimony used as the hardening alloy for the lead. This battery technology was able

to sustain the US Army until 1982. Some of the major disadvantages of this battery were

the time required to maintain the battery, and the need to add water on an almost constant

basis. Water was needed in a lea~-acid battery to charge the battery. Even though the

battery also creates water, the efficiency of the reaction was not high enough to generate

enough water for the battery to continue to operate properly. [4]

8

The 6TL battery was introduced in order to alleviate these primary concerns. Between

the 6TN and 6TL, there was no increase in storage capacity or battery life. Chemistry

advances were made to increase the efficiency of the

reactions which lessened the demand for water to a

great extent. The need to "top off' the battery only

occasionally with water became the common practice,

instead of every time the battery was cycled. Another

advantage was a decrease in the maintenance. With the

added efficiency of the battery, there were fewer

failures of individual cells. One of the driving forces Figure 1 - 6TL Battery

behind the added efficiency was the use ofless antimony in the cell. [4]

In 1996, the US Army began to transition into calcium instead of antimony to support the

lead in the battery. Antimony is a very toxic material, and along with the acid already

Figllre 2 -- 6TLFP Battery

present in the battery, the US Army wanted to utilize

a material that would be safer for the soldier. The

6TLFP was introduced, offering multiple advantages

to the 6TL predecessor. One major issue with the

6TL and 6TN battery was the self-discharge effect, or

the phenomenon of a battery to lose its charge while

not in use. Since the self-discharge rate was lower

than before, the 6TLFP had a greater shelf life and required less maintenance than the

6TL. One of the largest breakthroughs was the ability now for the battery to not require

any water to be added to the system, which further decreased the amount of maintenance

needed. [4]

In 1999, a minor advancement was made in battery

technology allowing the US Army to move to the

6TMF. The overall chemistry was the same as the

6TLFP; however the chemistry had been optimized

giving the 6TMF a greater overall capacity. This

9

increased capacity was thought at the time to give the soldier enough power to complete

missions. One of the few ways to tell the difference between the batteries was the color

of the casing: a 6TLFP was green, while the 6TMF is brown. Currently, this is the

official battery that is used by the US Army. [4]

In 2004, an update was made to the specification for the battery that the US Army uses in

the field. This update was made to allow competition to the archaic flooded lead acid

battery. This move forward has allowed not only the advances in lead acid technology,

but other battery chemistries to be able to compete for the US Army's business.

The leading candidate at this moment is the Valve-Regulated Lead Acid (VRLA) battery.

One of the major developments was the use of Absorbed Glass Mat (AGM) technology,

Figure 4 - VRLA Battery

which utilizes a thin fiberglass felt which holds the

electrolyte in place like a sponge. VRLA batteries are held at

a constant pressure (one to four psi), which promotes the

recombination process of the hydrogen and oxygen during

the charge cycle. This technique uses advances in

technology to give the soldier higher starting power, longer

shelf life, extended life, deep discharge capacity, lower life-cycle costs, among others.

Below is a table comparing the 6TMF with the VRLA:

10

Table 1 - 6TMFNRLA Comparison of Technologies [5]

6TMF VRLA

Voltage (V) 12 12

Cold Cranking Amps 650 240

Reserve Capacity (min) 200 240

Usable Reserve (DoD) 30% 70%

Shelf Life (months) 3 30

Battery Type Flooded Lead Acid Sealed, maintenance free

Cycle Life 235 350

Life (months) 13 48

Technology Lead Calcium flooded AGM, sealed recombinant

Internal Resistance (ohms) 0.009 0.0017

Transport Class Hazardous Non-spillable

Environmental Design Hazardous Non-hazardous

Weight (kg) 34 40

Size (I x w x h) (mm) 256 x 269 x 227 256 x 269 x 227

Even with the battery advancements made to date, there is still an issue with capacity and

depth of discharge (DoD). Below is a figure showing a "battery graveyard". This is a

recent picture of dead batteries from the field. When used by the soldier, they do not

always pay close attention to the DoD during usage. To complete their missions, a need

11

to utilize more capacity IS desired, thus taking the discharge down below its

recommended levels. This poses a major issue with logistics. A change needs to occur to

prevent this from happening.

SINGLE FUEL FORWARD

In earlier times, the US military had used a combination of fuels to power their arsenal.

Passenger cars used motor gasoline (MOGAS). Tactical and combat vehicles used diesel

fuel (DF-2). Aircraft used aviation fuel (Jet A, Jet A-I, and others). Ships used Diesel

Fuel-Marine (DF-M) and Bunker Fuels. Logistically, this was becoming unacceptable.

Conditions arose in which some vehicles became stranded due to the unavailability of the

correct fuel.·

The Single Fuel Forward policy was developed for all military vehicles, ground and air~

at least, for those not having a spark-ignition engine requiring gasoline. The fuel chosen

was JP-8. [6] JP-8 is a kerosene-like aviation fuel consisting of hundreds of

hydrocarbons ranging from CIO - C16. The hydrocarbons that make up JP-8 can' be

broken down into four distinct classes; paraffins (e.g. dodecane), naphthenes, aromatics,

and bicyclics (e.g. tetralin). It has a sulfur level comparable to on-highway DF2 in the

US of around 300 ppm, but the fuel quality can not be controlled in other countries,

where the military specification must be flexible to allow up to 3000 ppm sulfur. As a

side note, it was discovered that the flash point of this fuel was too low for ship board fire

safety. JP-5 was developed from JP-8 as a solution for this problem. Today, all vehicles

are commissioned to use JP-8 I JP-5 as the single fuel for the US military. This has

greatly reduced the logistical burden, as well as the cost of placing separate fuels into

required locations.

CLEAN FUELS INITIATIVE

The cost of fuel has been increasing rapidly, especially the past few years. The increase

is at least partially attributed to the cost of locating it and producing it from the wells. As

domestic production slows, the United States is becoming increasingly dependent on

imported sources of oil. Today, the US imports 57% of its oil requirements, and that is

12

expected to rise to greater than 70% by 2025. China, India, and Eastern Europe are all

expected to increase their demand by 54%. The developing world will increase their

demand for crude oil by 91% in a fierce competition with the Western nations. The

world consumption is expected to rise from 70 to 121 million barrels per day by that

time. [6]

Compounding these problems is the fact that no new refineries have been built in the

contiguous US in the last 25 years. Since 1985, 66 operating refineries have been shut

down, leaving 148 still operating. The capacity of the state of California alone has

decreased by 20% since 1985. Further, environmental laws and cultural considerations

impose restrictions for building new refineries, such as the culture of "Not-In-My-Back-

Yard". Of the remaining refineries, 88% of the capacity is located in eight states on the

Great Lakes and the East, Gulf, and West Coasts, including mega-refineries in Texas,

Philadelphia, New Jersey, Los Angeles, and Louisiana. These complexes are potentially

significant terrorist targets, not to mention being located in areas sensitive to the effects

of extreme weather conditions. Fuel prices skyrocketed after Hurricane Katrina· struck

New Orleans in 2005, disabling the Gulf oil refinery industry.

Unfortunately, there is a risk associated with maintaining the status quo. World oil

production will peak and reserves will fall. The US is currently on track to import 70%

of its crude and 25% of its refined products. A coordinated terrorist attack on US

refineries, ports, and marine terminals would place national security, public welfare,

economic security, and public confidence at risk.

In 2004, the US Department of Defense developed the Clean Fuels Initiative. The goal of

this initiative is to secure indigenous sources of energy, to use an environmentally

sensitive Fischer-Tropsch process to produce fuels, and to produce better fuels (no sulfur,

cleaner burning, bio-degradable), while using limited government funding and meeting

existing government mandates.

13

The primary source of fuel for this initiative is coal. The US has more coal in reserve

than any other country. In the central US alone, there are more than 250 billion tons of

coal, which is equivalent to 500 billion barrels of crude oil. The secondary source of fuel

is Petcoke. Petcoke is a carbonaceous residue produced by all refineries. Approximately

50 million tons per year are produced, equivalent to 100 million barrels of crude oil.

Converting these products into a useable fuel will reqUIre usmg an old, proven

technology from pre-WW II, the Fischer-Tropsch (F-T) process. This process was

developed by two Gennan scientists to convert indigenous coal into a liquid fuel, because

at that time, Gennany's ability to secure petroleum fuel was severely disabled due to the

war, The basic process is:

This is the conversion of a biomass fuel into carbon monoxide and hydrogen, the so-

called Synthesis gas reaction, which takes place in the presence of a catalyst. These

materials are then converted, in the presence of another catalyst, into long chain paraffins

through the reaction:

n CO + 2n H2 ---t (CH2)n + n H20

These paraffins are then converted into liquid fuels using cracking processes, which are

standard in the industry. After the war, the technology was shelved as it was deemed too

expensive to use, compared to inexpensive and plentiful petroleum. More recently,

during their apartheid embargo period, South Africa used this technology to help them

produce an F-T blending stock for jet fuels. Even today, the jet fuel at the Johannesburg

airport in South Africa contains a significant portion of Fischer-Tropsch fuel from their

iocal plant, SASOL.

Fischer-Tropsch fuels can be considered superior fuels for the following reasons:

a) Non-detectable levels of sulfur, aromatics, or metals

14

b) Higher cetane number than diesel fuel (> 74 vs. 45 - 50)

c) Very low toxicity and is bio-degradable

d) Produces lower emissions ( NOx, PM, and CO) and smoke when burned

e) Completely compatible with the existing fuel distribution infrastructure

f) Immiscible with water

g) Can be a source of hydrogen

Industry currently plans on building eleven F -T coal plants and nine Petcoke plants to

produce liquid fuel. This would supply 12% of the US energy needs, and will have a

direct impact on the price of crude oil. The eleven coal plants will supply the US military

with all of its worldwide requirements. The nine Petcoke plants will help current refiners

to convert a waste product into a clean blending stock, allowing them to produce an extra

900,000 bbl I day of fuel, along with an extra 10.5 GW of new power from the associated

co-gen plant.

REFOR.l\1ING FUEL

JP-8 is a difficult fuel to reform, but due to the requirements of the Single Fuel Forward

policy, it is a necessity for the US military. It comprises about 70% of the battlefield'

cargo stream. Highly efficient fuel cell APUs are being considered as a means to reduce.

fuel consumption in' the field, but would require the use of hydrogen created from

reformed JP-8.

There are three common reforming chemistry options: Partial Oxidation (POX), Steam

Reforming (SR), and Autothermal Reforming (ATR).[7]

Partial Oxidation (POX)

CnHm + 12 n O2 ~ n CO + Y2 m H2 IJ. HC8H18 = - 660 kJ/mol

Low H2 yield, exothermic reaction, mass transfer limited

Steam Reforming (SR)

CnHm + n H20 ~ n CO + (n + 12 m) H2 IJ. HC8Hl8 = + 1274 kJ/mol

High H2 yield, endothennic, heat transfer limited

Autothennal Reforming (ATR)

CnHm + X O2 + (2n - 2x) H20 ----+ n CO2 + (2n - 2x + 112 m) H2

Medium H2 yield, tunable heat duty [8]

15

A project was recently completed by the US Army to develop a fundamental

understanding of how major JP-8 components behave under ATR conditions. Previous

work had shown that one of the biggest hurdles to having a successful JP-8-fuelled fuel

cell was preventing the sulfur-laden JP-8 from poisoning the catalyst.[9] Specifically, it

was decided to compare the reforming performance of JP-8 paraffins (dodecane) to JP-8

bicyclic aromatics (tetralins). The catalyst system was 10 wt% Ni / CeO.7SZrO.2S02 (Ni /

CZO). This was chosen since it had previous success as a refonning catalyst for gasoline

sun-agates. The conclusions of this study were that:

1) Dodecane A TR was unstable for oxygen to carbon ratios less than 1.2

2) Tetralin ATR was unstable for 0 / C ratios less than 1.0

3) JP-8 surrogate compounds are much more difficult to reform than gasoline

surrogates

4) JP-8 refonnate was much more suitable for SOFC rather than PEMFC

The Army is currently initiating follow-up work. This study, "Assessment and

Development of Advanced Fuel Processing Properties," will provide an assessment of

promising fuel processing options, including the development of novel approaches for

reprocessing the Army's logistics fuel, JP-8.[10] There are three main objectives for this

program:

1) Develop reforming options for Solid Oxide Fuel Cell APUs

2) Investigate the feasibility of gas-to-liquid fuel processes for APUs and other

Army relevant applications

3) Develop technical approaches for overcoming barriers to the success of SOFC

APU demonstrations using JP-8 fuel

16

Objective number one involves investigation and research into two methods for

reforming JP-8 logistics fuel-direct decomposition and step-wise reforming, as well as

autothermal reforming for using JP-8 in SOFC APU applications.

The objective of the first sub-project is to demonstrate the feasibility of a simple, robust

and lightweight fuel processor that can be used with complex hydrocarbon fuels, such as

IP-8. The reformers that have been developed to date for mobile applications consist of

three reactors and a large number of heat exchangers. The result is a relatively heavy and

complex reformer that weighs significantly more than the fuel cell itself. This project will

demonstrate the feasibility of a reformer that is dramatically different, less complex, and

physically lighter. This approach to fuel processing is based on direct catalytic

decomposition of the hydrocarbon fuel into carbon and hydrogen. Using a catalyst, this

reaction is expected to occur around 600-800DC. A catalyst will be chosen on which the

carbon is deposited in the form of carbon fibers. The catalyst surface will be regenerated

by burning off the deposited carbon. The catalyst can also be regenerated by a mixture of

steam and oxygen, in which case it is possible to produce CO and additional Hz, in

addition to CO2 . In this regeneration approach, the catalyst bed will be heated to

,~lOOO°C. The heat deposited into the catalyst bed during regeneration will then supply

the energy needed to decompose the hydrocarbon fuel. The excess heat, carried away by

the hot gases from regeneration, will be used to preheat and vaporize the liquid fuel.

Overall, this process is conceptually similar to catalytic cracking used in petroleum

refineries and in the material production of carbon fibers.

When estimating the size of the reactors and catalysts for this type of a fuel processor and

comparing it to the three-component type (reforming, water-gas shift and preferential

oxidation), it will weigh significantly less than the classical fuel processor because it

eliminates the heavy water-gas shift and the preferential oxidation reactors. The number

of heat exchangers will also be reduced to only one. The construction materials will be

simpler and more robust, because reactors of this type are used in automotive exhaust

emissions aftertreatment, such as ceramic honeycomb reactors. There will be no need for

17

extra heat exchangers because the catalyst beds will serve as the heat sinks (regeneration)

and sources (decomposition).

The second sub-objective is to study the effect of complex feed mixtures on the catalytic

activity and durability of fuel reformers. For transportation fuel cell applications, it can

be critical to be able to reform the current hydrocarbon fuels into hydrogen for onboard

demands. Depending on the application, there are different specifications for the required

hydrogen purity. Hydrogen for a low-temperature fuel ceH, such as the polymer

electrolyte fuel cell (PEMFC), must contain less than ten ppm of CO to eliminate anode

poisoning. However, the purity requirements for a high-temperature, such as SOFC, are

much less stringent, because the high temperature eliminates CO site poisoning concerns.

Little is known of the effects of hydrocarbon mixtures on reforming catalysts' activity

and durahility. Diesel and JP-8 are both quite different from the commonly tested iso-

octane and methane. Not only do they consist of larger and more complex hydrocarbons,

but they also contain fuel additives that could have .detrimental effects on both catalyst

durability and refonning chemistry.

Equally important to understanding the effect of complex feeds on state-of-the-art

reforming catalysts, it is also necessary to develop novel catalysts that remain active over

a wide variety of fuels. Preliminary research has shown that catalysts with ion

conducting supports show an improved activity over traditionally supported catalysts. In

this project, supports of interest may include, yttria-stabilized zirconia (YSZ), zirconia-

doped ceria (CeZr02), gadolinia-doped ceria (GDC), and, Gd2Ti20 7. These materials

display oxygen mobility of some form, mid are the subject of significant research. Many

of these support materials are key components in SOFC construction. For example, YSZ

is often used as both the electrolyte and the anode support material in a SOFC. While

SOFCs have the ability to reform hydrocarbons directly, in practice, an indirect reformer

is often placed upstream of the SOFC due to thermal management issues. Direct·

reforming would be ideal for SOFCs since it would greatly simplify the SOFC system. If

novel catalysts could be developed that are functional for fuel reforming and have

18

attractive electrical properties, it IS conceivable that they could be used for direct

reforming inside an SaFe.

Objective number two is investigate the feasibility of gas-to-liquid fuel processes for

APUs.

The first sub-objective is to conduct fundamental research and process development for

the production of a reformer-friendly designer fuel, via advanced Fischer-Tropsch

synthesis. The Fischer-Tropsch synthesis permits the conversion of carbon monoxide

and hydrogen into a variety of organic compounds, as illustrated below:

An intriguing possibility is to use the F-T synthesis for production of a designer fuel,

meeting exact specifications. The product distribution obtained is dictated by Anderson-

Schulz-Flory polymerization kinetics, reSUlting in a product mix that is far from ideal for

purposes of transportation fuels. Consequently, the products emanating from an F-T .

plant require substantial downstream processing to meet transportation fuel

specifications. On the other hand, the process is attractive insofar as the resulting

hydrocarbon mix is free of sulfur. This is of great importance for use of a fuel in on-

board fuel processors for SOFC auxiliary power units.

There has been considerable effort, both in industrial as well as academic laboratories, to

understand the factors contributing to positive or negative deviations from the product

distribution of the Fischer-Tropsch synthesis by the Anderson-Schulz-Flory

polymerization kinetics. To create this "designer" fuel, it is important to learn how to

manipulate the product distribution. To date, there have been some observations of small

deviations for the theoretically predicted product distribution, as well as illustrating that

the paraffin/olefin ratio increases exponentially with increasing carbon number.[13]

Several explanations for this phenomenon have been offered in the literature. One states

that the catalyst sites produce olefins as primary products, and that olefins formed are re-

19

adsorbed on different sites on the catalyst, where they undergo hydrogenation.[14]

Others have suggested that the increasing paraffin/olefin ratio as a function of

hydrocarbon chain length is due to transport rate differences[ 15] or differences III

solubility.[16]

The first strategy consists of manipulating the surface composition of advanced Fischer-

Tropsch catalyst formulations on an atomic scale. Recent advances in combinatorial

synthesis methods and high-throughput catalyst screening methods, along with advances

in computational methods, will be explored for rapid optimization of multi-functional

Fischer-Tropsch catalysts, capable of creating product distributions that are "reformer-

friendly".

The second strategy involves novel catalyst formulations, where catalytically active

metals will be supported on refractory support materials that exhibit fast ionic

condlictivities. Several fast ion-conducting solids that allow transport of various mobile

. . h H+ r'+ N + A ++ C ++ Pb+2 F- d 0-2 '1 bi lOn speCIes, sue as , .. _..1, a, g, U,.., "an, are aVal a e

(Agrawal, 1999).[17] These crystalline solid electrolytes provide a rigid· lattice

framework throughout the bulk structure, which contains channels with unique and

specific structural features in which one of the ionic species of the solid, such as oxygen

ions, can migrate. Ionic transport involves site-to-site hopping along these conductive

chamlels. There is reason to believe that catalyst supports with high oxide-ion mobility

will permit the fine-tuning of Fischer-Tropsch product distributions, analogous to what

can be accomplished with doping. The primary difference, however, is that

electrochemical promotion is tunable, so that the promoter effect can be optimized to

yield a "reformer-friendly" fuel. The revolutionary potential benefit of using solid ionic

conductors in Fischer-Tropsch catalysis lies in the electrochemical promotion of activity.

By applying a voltage across a monolithic support, ions, such as sodium, can be pumped

to or from the working catalytic surface, producing changes in activitJ:. and product

selectivity. This may lead to "tunable doping" where different product distributions can

be achieved by changing the applied voltage.

20

The third strategy that will be evaluated is the use of modular reactor network systems·

with localized heating and multi-port, distributed feed. The product distribution leaving

one reactor can be adjusted by co-feeding a second reactant into the input stream into that

same reactor. An ideal process would achieve lOO% selectivity towards desired products,

thereby eliminating the need for product separation and purification. An important step

in this direction is to move towards "smart" catalytic systems that are capable to respond

to agile, localized control strategies. Utilizing "smart" catalytic modules instead of bulk

catalysts opens exciting opportunities to explore agile, localized control strategies.

The second sub-objective is to improve the reforming characteristics of JP-8 by selective

removal of aromatics. The difficulty in reforming JP-8 fuel for the production of H2 is

mainly due to its higher aromatic and sulfur contents. The presence of various additives

is also known to have deleterious effects on the performance of steam reforming

catalysts. Significant research effort is being directed to the development of suitable

adsorbents to remove sulfur from JP-8 but little attention is being paid to the issue of

aromatics. Those present in JP-8 are mostly alkyl-substituted naphthalenes and

polyaromatic hydrocarbons[l9]. Naphthalenes and other polycyclic aromatics have

higher tendency toward coking than mono-aromatic compounds and, therefore, the

currently available methane and gasoline steam reforming catalysts are not suitable for

the processing of JP-8. In addition to developing better reforming catalysts for handling

these aromatics[2l], possible ways of pre-treating JP-8 should also be considered. These

pretreatment processes can be based either on physical removal of aromatics or on

catalytic conversion of these refractory compounds to the more easily reformed ones.

The aromatic compounds present in JP-8 fuel can be removed through solvent extraction

processes, a common practice in refineries for removing aromatics from kerosene.

Research is required to develop a suitable compact extractor for JP-8 applications. If

properly developed, such a pre-treatment step could also partially remove the sulfur

compounds as well as the additives that are present.

21

The transformation of polyaromatic hydrocarbons to mono aromatics and paraffins have

recently created significant research interest for reducing aromatic content in diesel. The

reactions generally take place under elevated pressure (about 100 bar) and temperature

(300-400°C) in presence of H2 over catalysts having both hydrogenation and cracking

capabilities. Researchers are also studying catalytic cracking as a pretreatment step for

converting the polyaromatics to monoaromatics. Further studies are required to develop

improved versions of these processes or newer processes based on novel concepts for

converting polycyclic aromatics to monoaromatics.

Objective number three is to develop a technical approach for overcoming barriers to the

success of SOFC APU demonstrations using JP-8 fuel.

Examples of important problems associated with SOFCs are:

.. {a) Carbon coking - For a hydrocarbon fuel, the state-of-the-art NilYSZ (Ni .

. . supported on yttria-stabilized-zirconia) anode material gets poisoned by graphitic

carbon depositions, which limit its catalytic performance. Novel anode materials

need to be developed that are more carbon-tolerant than Ni.

(b} Over-potential losses - Almost any SOFC design has an over-potential

problem. Activation losses are associated with slow inherent rates of catalytic·

electrochemical reactions. Novel anode materials that are inherently more active·

catalysts for electro-chemical oxidation reactions will be identified.

(c) Sulfur poisoning - Almost any hydrocarbon fuels, such as methane, diesel, JP- ,

8) and gasoline, have a certain content of sulfur containing molecules. These

sulfur compounds tend to poison Ni anode catalysts and dramatically reduce their

effectiveness. It is important to formulate anode catalysts that are more sulfur-

tolerant than Ni anodes.

The first sub-objective will approach this issue by developing advanced catalysts for

SOFC anodes and cathodes using density functional theory and laboratory experiments.

The quality and economical feasibility of SOFCs will ultimately depend on efficient,

robust, and affordable fuel cell electrode materials. There is a very limited understanding

22

of the molecular level chemical processes that govern electrode performances. This sub-

objective proposes to use quantum Density Functional Theory (DFT) calculations and

well-controlled surface science experiments to study molecular level mechanisms of the

electrochemical reactions that take place on the SOFC anode.

The molecular-level mechanistic information will be used to search computationally,

using quantum DFT calculations, for anode materials that are more resilient to carbon

coking and sulfur poisoning, and that have lower over-potential losses as compared to Ni

anodes. Any identified metal alloy will be synthesized and tested using an experimental

setup designed for fuel cell testing. Subsequent optimization of these materials will then

occur, specifically with respect to JP-8.

The second sub-objective is to make a determination of anode catalyst deactivation

mechanisms during exposure to lP-8 reformate and identify counter-measures. SOFCs

have several advantages over other fuel cells in APU applications. Internal reforming of

hydrocarbon fuels in the cell anode compartment possible with an efficiency of fOlty to

si~(ty percent, eliminating the need for external refonning because SOFCs operate at an

devated temperature higher than 600°C. One problem with using SOFCs is that during

internal reforming of hydrocarbons, coke formation is possible, which has a negative

impact on the anode catalyst life.

The purpose of this sub-objective is to conduct research into the possibility of internally

reforming JP-8 fuel in SOFCs. A literature review indicates that current research is

focused on the internal reforming of methane in SOFCs but there is a lack of knowledge

on intemal reformation of other hydrocarbon fuels, such as JP-8. The main challenge

with methane internal reforming is coke formation. The Army is evaluating the

feasibility of indirect internal reforming as a way of minimizing coke formation during

the internal reforming of methane in an SOFe. This concept was proven to have

problems, due to the mismatch between the thermal loads created by the rapid

endothermic steam reforming reaction and the electrochemical reactions.

23

This sub-objective will be focused on attempting to optimize the performance of the

anode material as an alternative to internal reforming of JP-8, without catalyst

deactivation. The literature on methane reforming will be used as a guide to JP-8

reforming. The feasibility of using indirect internal reforming will be evaluated, as it is a

way of internally reforming methane without causing coke deposition in the anode

compartment of the SOFC. This analysis will lead to optimizing the performance of the

anode catalyst as an alternative to internally reforming JP~8 without anode deactivation.

If this research shows that internal reforming is not feasible for JP-8 reforming in SOFCs,

then external reforming will be considered as a secondary option.

There is still much to be learned regarding reforming JP-8 into a useable fuel for SOFCs.

The Army intends to pursue these multiple options for the advancement of converting JP-

8, the logistical Single Fuel Forward, into a fuel that is useful for SOFCs.

MILITARY JP-8 REFORMER

Per the Single Fuel Fonvard policy, there may be a future need to make JP·8 and separate

out ,rh~ hydrogen. As stated earlier, JP··8 is a sulfur-rich, kerosene-like fuel that· lS

difficult to reform. The three driving forces for this change include:

1. With its current capabilities and equipment the US Army cannot meet the

objective of "Silent Watch".

2. With the continued advancement of vehicle electrical and electronic power

requirements, vehicles are reaching critical levels to where decisions are

made as to what electronics are operating at any given time to complete

the mission.

3. Currently there 1S no power management system available for use to

manage the power demand. [23]

24

Table 2 - Comparison of Usage of JP-8 and Cost per 12.5 kWh [23]

$ Required JP-8

/12.5 kWh gallons

Abrams $2470 99

Bradley $390 15.6

Stryker $312 12.5

Fuel Cell System $25 1

Table 3 - Comparison Usage of Cost to Charge Battery [23]

I Fuel Usage (gal/hr) Total Battery Gallons to Cost to I

Basic Tact. Possible Power charge at charge

Idle Idle Power (kWh) Tact. Idle batteries

(kWh) (2 hr)

brams 10.784 14.23 9 3.6 28.46 $711.48 ' -----t

r~ley £.899 1.498 6 2.4 3.00 $74.89

1°.599 --

3 1.2 1.20 $29.96 tryker 0.374 --

Table 41 - Scenario: 100 Charge Cycles (JP-8 reformer and fuel cell APU $28,500.

estimate 2010) [23]

~ost 100 Charge Cycles Savings / gallons conserved Payback Time

I 12.5 kWh/cycle # Charge cycles

'Abrams $247,000 $244,500 9789 gal 12

Bradley $39,000 $36,500 1460 gal 78

[strYker $31,200 $28,700 1148 gal 99

L~'uel Cell $2500 I

Other benefits, besides cost savings, are a reduced engine operation/improve main engine

life, reduced engine maintenance, maintenance schedules and logistics support, and

reduced battery weight and space claim. Currently, this program is in the process of

detennining how many contracts to award from the proposals submitted to the US Army

T ARDEC Broad Agency Announcement (BAA).

25

HYDROGENICS MULTI-SERVICE REGENERATIVE ELECTROLYZER FUEL CELL

The Hydrogenics Multi-Service Regenerative Electrolyzer Fuel Cell (MREF) IS a

potential alternative for currently fielded APUs. Unlike some of the other initiatives,

which focus on utilizing JP-8 as a source of hydrogen, the MREF system creates

hydrogen from water through electrolysis, which can then be used in the fuel cell.

This is an on-going project that began back in 1999 with a paper study to determine the

path forward for the Army. It is mandatory to have sufficient power for continuous Silent

Watch for a specified amount of time. Silent Watch is a situation where the main engine

is shutdown and the crew still needs to operate electronics and other equipment without

being detected. Other application requirements include a unit that is quiet, cool, efficient,

and that will supply enough power to complete the missions. The result of the study

determined that a 5-10 kW Proton Exchange Membrane (PEM) Fuel Cell system would

address the all-inclusive needs of the soldier.[24]

The second part of the program was to demonstrate a viable technology. In 2001, an

agreement was signed and a contract was placf.:d for this unit; the agreement included the

US and Canadian militaries. The goal

was to create a unit that would supply 3

kW of power, with the ability to run up

to 5 kW for short periods of time, have

10 kWh of hydrogen storage, and

operate at 28 volts DC. This unit was to

be tested in a laboratory environment

and be abl~ to charge off a standard 300

A alternator. The vehicle that was

chosen to support this effort was Stryker,

the Light Armored Vehicle (LA V). To

store the hydrogen, a metal hydride canister was chosen. Although compressed gaseous

hydrogen was a proven storage method, the program went a step further to pursue the less

26

developed, and higher risk, high-pressure metal hydride storage.

demonstration and evaluation occurred in 2002 at the US Army TACOM.[24]

Successful

The next phase of the project was a vehicle

demonstration program. Multiple branches of the

US and Canadian military were involved in the

design process; there were increasingly more

power and storage demands put on the unit. The

new power requirement was 7 kW of power, with

9 kW for short durations, and 30 kWh of storage

was specified. The three main components of this

system are the fuel cell stack, the electrolyzer, and the hydrogen storage.

Below is a listing of the technical specifications of the fuel cell stack:

Table 5 -- MREF Stack Specifications[24] -

Dimensions cm 66.6 x 43.6 x 22.8 ._-- -

Mass kg 60

I Efficiency @ 50 amp % 51 Operating Lifetime hrs 1000

I

Net Rated Electrical Power kW 7

I Operating Current Range Amp 0-140

I Beginning of Life Voltage V 53-75

I Range ~Time from Off Mode to Idle sec < 15

Time from Idle to 7 kW sec

27

Below is a listing of the technical specifications for the electrolyzer:

Table 6 - MREF Electrolyzer Specifications[24]

Wattage Demand at 28 VDC 4-6kW

Hydrogen generation 20 slpm (2.5 kg/day)

Supply gas pressure 100 psig

Start-up time 1 minute

Time to Max Capacity ~inutes Purity of generated hydrogen 99.9~------

Effective Electrolyzer on-time

Equivalent hydrogen produced

Figure 9 - MREF Phase III: Electrolyzer

System on-time --_.

Operating voltage ._-------

Nominal current

-Number of on/off cycles ----_.

The hydrogen storage is in three separate metal hydride

containers. Each canister has dimensions of 14.0 cm

diameter and 16.5 cm in length, a weight of 110 kg, and

a hydrogen storage capacity of about 13 kWh. Even

though these canisters have quite a bit of storage

·.;apacity, they arc very heavy. Unfortunately, the

military needs more production, with less volume and

less weight. [24]

3800 hrs

3200 standard m j (282

kg) ----

,-6000 hrs

28V _. --

150 A ~ 15 slpm of

hydrogen _.-

> 500 . ____ .J

Currently, testing is being done on this system to determine whether or not it IS

compatible with the electronic systems during a "Silent Watch" profile.

28

The next phase of the program is being developed. The initial goal was to demonstrate

the ability of the system to operate in a military environment, but the system will have to

be redesigned and integrated into a vehicle. Multiple tests will then be conducted to

verify the durability and ruggedness of the system. At this time the decision has not been

made whether gaseous hydrogen will be utilized, or if metal hydride will still be the

hydrogen storage.

This system has several technological advantages for military applications. Most

importantly, extending the period of continuous "Silent Watch" is a necessity. Along

with the advancement of technology, there is a greater demand for power, and the MREF

will be able to supply the power demands. Another advantage is the system's ability to

use water as the source of hydrogen, as opposed to JP-8 or other fossil fuel. With the

ability to create hydrogen, store hydrogen, and produce power, the MREF has the

advantage to "plug and play" on multiple ve~icle fomlats without major redesigns. Since

both the US and Canada are interested in reducing their dependence on tossil fuel, this

collaborative effort is an appropriate stepping stone to a hydrogen economy.[24]

However with all emerging technologies, there are also disadvantages. Although water is

a great and abundant resource, the greatest drawback is its relatively near-ambient

freezing point. At Qoe, potential system damage needs to be minimized. In addition, for

any dynamic condition, there are issues for the system. Vibration, dust, puncture, as well

as temperature are all requirements that will be challenging for the MREF to pass.

Another disadvantage of this system is the amount of space the MREF requires. In the

current phase, metal hydride tanks are still utilized; these comprise a majority of the

volume, as well as the weight. Optimization will reduce this footprint. Logistic

optimization is another issue. Since water is being used as a fuel source, additional

demands will arise for the water. There will also be a need for electrolyzers to operate at

base camps. Further, there is a question as to whether there will be enough hydrogen

produced inside the mobile units. Another issue will be whether the quality of water

needed to operate this system is critical, or else will the system's "water requirements"

require a special water supply in the battlefield.

29

DELPHI SOFe APU wi REFORMER

Figure 11 - Delphi SOFe APU with integrated fuel processor. The picture on the right illustrates the volume compared to a

typical battery. [26J

As shuwnabove in Figure 11, Delphi Corporation has been developing a five kilowatt

solid (}xide fuel cell (SOFC) APU, with integrated fuel processor, for use in military and

commercial heavy-duty vehicle applications. A modeling effort, fUHded by the US Anny

National Automotive Center, was conducted to determine the estimated performance of

this APU on a Bradley M2A3 Diesellnfantry Fighting Vehicle (IFV). The vehicle holds

a crew of three (commander, driver, and gunner) and up to six infantrymen. The main

engine is a diesel Cummins VTA-903T 447kW (600 BHP), and there are six 12V and one

24V batteries, as shown in Figure 12 and Figure 13 below. In addition to poor engine

fuel economy at idle, the battlefield identification risks are increased for the soldiers in

the: vehicle, due to the main engine's heat rejection and visible exhaust smoke. [25]

Mechanical Drive Shafts

__ Drive Shaft to Track

Figure 12 - Schematic layout of the current BFV powcrtrain.

Figur'! 13 - Schematic layout ofthe current BFV battery and generator configuration.

30

During Silent Watch operation, III which an electrical load of 85A and 2.3kW are

required, the fuel economy observed by using an APU was improved by 86% in

comparison with the main engine, despite the efficiency loss of having a fuel processor.

This extended the continuous Silent Watch capability from the baseline tive days, when

relying on the main engine, to over 36 days, when relying on the SOFe APu. As shown

in Figure 14 below, the Bradley's electrical requirements are escalating quickly in power

demand, and extending the Silent Watch capability with a more efficient APU is highly

desirable. There were incremental vehicle efficiency improvements when modeling the

engine hull fan and water pump, in which those electric engine accessories were powered

31

by the APU. Removing parasitic losses on an engine translates to more engine power

being available for purely propulsion purposes; this provided marginal improvement in

acceleration from 0-30 MPH. However, in the situation of the vehicle physically moving

at a maximum speed of 2600 RPM and 75% load, the fuel cell did not prove to have a

beneficial impact on fuel economy or vehicle performance. [25]

!

I 1

Bradley Electrical Generating Capacity

j -------1 I ,--------,----'-----r--- .----,---

Figure 14 -- Bradley electrical power requirements increasing over time (provided by the US Army T ACOM Bradley PM

office).

FREIGHTLINER TRACTOR WITH BALLARD PEM APU AND METHANOL REFORMER

The Army Transformation Initiative has been providing direction to the organization to

develop lighter combat vehicles, and therefore lightening its logistic infrastructure. As

General Kern stated at the 2003 SAE World Congress, approximately two-thirds of the

military's vehicles in the battlefield deliver fuel to the other one-third, and that

approximately two-thirds of the fuel consumed in the battlefield is in the transport of that

fuel, as illustrated below in Figure 15. There is also a rough estimate that the military

spends $100-500 in logistics cost for every gallon of fuel transported to the battlefield, so

substantial improvements to fleet fuel economy are an absolute necessity. Reduction of

fuel consumption can be achieved by minimizing the delivery, dispensing, and storage

infrastructure needed through fuel economy improvements, as well as being able to

reduce the combat assets needed to protect the fuel re-supply infrastructure. [3] [25]

Combat Element

---.1 I.-

+- Logistics Head---.

Cumulative Miles

I

33

Figure 16 - The Freightliner Class 8 tractor with the Ballard integrated APU.

Risk analysis was performed regarding vehicle safety in the case of fuclleakage, and the

safest APU location was determined to be behind the cab on the frame rails, as shown in

Figure 17. The module itself is shown in Figure 18. [2] [3]

Figure 17 - APU installed on the side frame rails of the tractor. The fuel tank is mounted on the opposite side of the truck.

Figure 18 - The Ballard module assembly. The stack is on the top, the air system is on the bottom right, and the fuel processor

is on the left.

34

The liquid methanol was actually a mixture of MeOH and water, and this was injected

into the fuel processor. It was vaporized prior to the reformer, and then converted by a

catalyzed reaction into a hydrogen-rich gas. The H2 concentration at this point was over

50% by volume and that of CO was

35

System efficiency, including the fuel processor, was found to be consistently greater than

30%. To meet the extremely fast transient response times necessary for a transportation

application, the fuel cell system was hybridized with a battery, providing a dynamic

response time of less than one second. As illustrated in Figure 20 below, the APU

provided a 16-68% improvement in fuel economy over idling the main engine. [2] [3]

Load

Figure :W -Data generated by the DOE NRF.L. ADVISOR model, c"Imparing the idling of a main diesei engine to this APU

c~ntigur,uion.

SUNLINE TRACTOR WITH HYDROGEN-FUELLED HYDROGENICS PEM APl) AND

VEHICLE ELECTRIFICATION

Figure 21 - The SunLine tractor after its cross-country trek from southern California to Washington, DC, in 2005.

36

The Delphi and Ballard programs outlined above show that there are fuel economy

benefits obtained during Silent Watch mode in utilizing a fuel cell auxiliary power unit

over main engine idling. To further ~mploy those benefits when the vehicle is actually

moving, the Army NAC wanted to determine other ways to gainfully employ a fuel cell

APu. Otherwise, the fuel cell system (FCS) was "deadweight" on a moving vehicle. A

program between SunLine Transit Agency, Southwest Research Institute, and the US

Army NAC tackled this exact issue, and they are showing that vehicle electrification is a

very complementary technological path to fuel cell APUs, utilizing the generated

electrical power without the typical mechanical-to-electrical conversion losses, not to

mention the primitive optimization. [27] [28] [29] [30]

Electrification offloads parasitic loads on the engine. This means removing an accessory,

such as the mechanical engine radiator fan, and replacing it with a comparable electrical

system. The accessory is longer tied to engine speed and load. Unnecessary excess

accessory -power demand is no lo'llger going to waste, because of system control

optimization through modulation. Heavy-duty vehicle systems that are electrified exist ,

. today include the engine fan, water pump, oil pump, air compressor, and cabin air

conditioning system. [27] [28] [29] [30]

Although in its infancy stages of development for commercial vehicles, waste heat

recovery can be another complementary electrification technology. By using electric

turbo compounding and an integrated motor starter generator (ISG), waste heat can be

recovered and converted into even more electrical power.

System electritication can be applied to anything from addressing commercial vehicle

idling reduction to enhancing Silent Watch capability, not to mention fuel economy

improvement. In controlled dynamometer testing of the SunLine vehicle ("pre-

electrified" vs. the current configuration), a 14% improvement in diesel fuel economy

was attained. Even when including the hydrogen consumption, the fuel economy benefit

is still substantial, and is the primary reason that the heavy-duty vehicle industry is

37

quickly integrating this electrification technology into an industry standard. [27] [28] [29]

[30]

Another vital role of electrification is as an enabling technology for fuel cells. Although

fuel cell supporters would like to see them "replace" internal combustion engines in

every application now, this will decidedly not happen immediately for the foreseeable

future, primarily due to the lack of hydrogen infrastructure, as well as the exorbitant cost

of the fuel cell systems themselves. Introducing smaller, less costly fuel cell APUs can

help develop hydrogen refueling capability in a more realistic manner and provide a

smoother transition.

The SunLine vehicle currently has two 10 kW Hydrogenics PEM fuel cells. The ultimate

goal of the program is to have a liquid-fuelled fuel cell APU, but progress is very slow in

developing a sulfur-tolerant fuel processor, mostly due to the decision by the auto

companies and DOE a few years ago to pursue on-board hydrogen storage instead of on-

vehicle fuel reforming. The vehicle has three 5000 psi Dynetek compressed gaseous

hydrogen storage tanks, for a total of 5 kg of hydrogen storage. The electrified system

components include an engine radiator fan, a water pump, air compressor, and air

conditioning system. These are all 42V systems, due to the perceived evolution to that

voltage a few years ago in the automotive industry. These electric accessories are all

powered by the fuel cell APU, but there is also a 42V alternator set up to provide back-up

power in case of fuel cell system failure (or if the vehicle runs out of hydrogen on the

road). [27] [28] [29] [30]

CONCLUSIONS

Although the US military is still many years away from implementing fuel cells in the

battlefield, they are working diligently to move the technology forward for dual-use

applications in commercial and military vehicles. The most critical issue is

desulfurization of jet fuel, since fuel cells are sulfur-intolerant and fuel quality can not be

controlled outside of the boundaries of the US. Demonstrations of hydrogen-fuelled fuel

38

cell APUs are a stepping-stone to liquid-fuelled fuel cell APUs, when refonnation

reaches the level of maturity required for system integration and demonstration.

39

REFERENCES

[1] SAE 2001-01-2792 "US Army Strategy for Utilizing Fuel Cells as Auxiliary Power

Units," by Herb Dobbs Jf. (US Army RDECOM TARDEC NAC), Erik Kallio (US

Army RDECOM TARDEC NAC, and James Pechacek (US Army RDECOM

T ARDEC NAC).

[2] SAE 2003-01-0266 "Recent Results on Liquid-Fuelled APU for Truck Application,"

by Massimo Venturi (Ballard), Erik Kallio (US Army RDECOM T ARDEC NAC),

Scott Smith (Freightliner), J. Baker (University of Alabama), and P. Dhand

(University of Alabama).

[3] SAE 2003-01-0267 "Synthetic Hydrocarbon Fuel for APU Application: The Fuel

Processor System," by Massimo Venturi (Ballard), Detlef zur Megede (Ballard),

Berthold Keppeler (Ballard), Herb Dobbs Jf. (US Army RDECOM TARDEC NAC),

Erik Kallio (US Army RDECOM T ARDEC NAC).

[4] Interview with Clarence Woodard, TARDEC Engineer, 4 October 2004.

[51 j1ttp:/jy..rww.milbatteries.com/hawkeribrochure:html

[6] US Energy Information Administration

[7] University of Michigan, Dept. of Chemical Engineering, B. Gould, A Tadd, J ..

Schwank

[8] University of Michigan, Dept. of Chemical Engineering, B. Gould, A Tadd, J.

Schwank

[91 J. Hu, Y. Wang, D. VanderWiel, C. Chin, D. Palo, R. Rozmiarek, R. Dagle, J. Cao, J.

Holladay, E. Baker Chemical Engineering Journal, 4049, (2002), 1-6

[10] University of Michigan, Advanced Energy and Manufacturing Technology, Phase HI

project

[11] R. Fischer and H. Tropsch, Brennstoff-Chemie 4 (1923) 276.

[12] G. A. Olah and A. Molnor," Hydrocarbon Chemistry", John Wiley &Sons, Inc~,·

pp.66 (1995).

[13] E. Iglesia, So C. Reyes, and S. L. Soled, in "Computer-Aided Design of Catalysts"

(E.R. Beckes and C.J. Pereira, eds., Marcel Dekker, New York, pp. 199-259 (1993).

[14] H. Schulz and H. Gokcebay, in "Catalysis of Organic Reactions", J.R. Kosak, Ed.,

Marcel Dekker, New York, pp. 153-169 (1984).

[15] E. Iglesia, S. C. Reyes, R.J. Madon, Journal of Catalysis 129 (1991), 238.

[16] E. W. Kuipers, C. Scheper, lH. Wilson, LH. Vinkernburg, and H. Oosterbeek,

Journal of Catalysisl 58 (1996) 288.

[17] R. C. Agrawal, and R.K.Gupta, Journal of Materials Science 34, (1999),1131.

[18] Villermaux, J. "Future Challenges in Chemical Engineering Research", Trans.

IchemE, 73, p. 105 (1995).

[19] T.J. Campbell, A.H. Shaaban, F.H. Holcomb, R. Salvani, M.J. Binder, J Power

Sources, 2004, 129,81.

40

[20] R. Coll, J. Salvado, X. Farriol, D. Montane, Fuel Proc. Technol., 2001, 74, 19; and

X. Wang, R.j. Gorte, Appl. Catal. A: Gen., 2002, 224, 209.

[21] B.N. Bangala, N. Abatzoglou, E. Chornet, AIChE J, 1998,44(4),927.

[22] M. Freemantle, "Microprocessing on a large scale", Chemical &Engineering News,

Oct.11, 2004, pp. 39-43.

[23] Presentation from 2005 Power and Energy Symposium "Fuel Reformation", Mike

Berels, TARDEC Engineer.

[24] Presentation from 2005 NATIBO Conference "MREF", Chris Spangler,TARDEC

Engineer, June 2005.

[25] SAE 2004-01-1586 "Logistics and Capability Implications of a Bradley Fighting

Vehicle with a Fuel Cell Auxiliary Power Unit," by Joseph Conover (Delphi / EDS),

Harry Husted (Delphi), John MacBain (Delphi), and Heather McKee (US Anny

RDECOM TARDEC NAC). Presented at the 2004 SAE World Congress.

[26] http://www.delphi.com/pdf/ppdienergy/sofc auto.pdf

[27] SAE 2004-01-1478 "42-Volt Electric Air Conditioning System Commissioning and

Control for a Class-8 Tractor," by Bapiraju Surampudi (SwRI), Mark Walls (SwRI),

Joe Redfield (SwRI), Alan Montemayor (SwRI), Chips Ingold (Modine), Jim Abela

(Masterflux).

[28] SAE 2005-01-0016 "Electrification and Integration of Accessories on a Class-8

Tractor," by Bapiraju Surampudi (SwRI), Joe Redfield (SwRI), Gustavo Ray (SwRI),

Alan Montemayor (SwRI), Mark Walls (SwRI), Heather McKee (US Army

RDECOM T ARDEC NAC), Tommy Edwards (SunLine Transit Agency), Mike

Lasecki (EMP).

[29] SAE 2006-TBD "Accessory Electrification in Class 8 Tractors," by Joe Redfield

(SwRI), Bapiraju Surampudi (SwRI), Gustavo Ray (SwRI), Alan Montemayor

(SwRI), Heather McKee (US Army RDECOM T ARDEC NAC), Tommy Edwards

(SunLine Transit Agency), Mike Lasecki (EMP).

[30] http://www.swri.org/sunline

41

)5363 PP

Feasibility of Fuel Cell APUs· for Automotive Applications

Chris Spangler Eric Sattler

Heather McKee

Wayne State University AET511 0 Professor ,Simon Ng

1217105

1

Overview • Vehicle batteries • Single Fuel Forward Policy • FC APU wi IP-8 reformer • FC APU benefits • Demonstration programs

• Delphi • Hydrogenics MREF • SunLine vehicle • Freightliner vehicle

• Clean Fuels Initiative • FT fuel

• JP-8 reforming

2

Battery History 6TN/6TL (1950/1982)

.Developed in 1950s for need ~.Large water demand

.High maintenance cost

6TLFP (1996)

.Change from calcium to anitmony

I ·Less self-discharge rate

3

Battery History . ,-- -------~~-.--------.. ------------~-I

• 6TMF (1999) i -Optimized chemistry

-Greater capacity

_I -Case colo_r_c ___ _ I -------

VRLA (2004)

-Absorbed Glass Mat

-Sealed -[Jeep discharge tolerant ____ ~J

4

Single Fuel Forward • Previous times, various fuels

• Diesel, gasoline, jet fuels

• Incorrect fuel left vehicles stranded!

• Caused logistical nightmare!

• All deployed vehicles "converted" • New fuel is JP-8

• Works for compression ignition and turbine engines

• Also suitable for ships as JP-5 • Higher flash point for fire safety

6

-,'

Single Fuel Forward ------

• JP-8 constituents • Hydrocarbons ranging from CIO to CI6

• Paraffins, napthalenes, aromatics, bi-cyclics

• Spec limit for sulfur up to 3000 ppm!

• Intolerant of reformation

-- .~ ... 7

Military JP-8 Reformer Challenges r'/

~

-Not meeting objective of "Silent Watch". I

-Electronic operation trade-ofts

-No power .managern.ent system available to manage the demand

8

Benefits of a Fe APU • Commercial vehicles - idling reduction

• Quieter • Less/no exhaust emissions • Less vibration

"'7 c ,-F

• Military vehicles - Silent Watch /,/' .,r • Quieter • Less/no exhaust emissions • Less vibration • Increase continuous operation • Reduced thermal signature

10

MREF Program I . .

I Phase 11- Concept Demonstrator

• 3 kW of Continuous Power

• 5 kW Peak Power

• 20 kW-hrs ofH2 storage

• 28 VDC format

• In a Laboratory Environment

• Std. 300A alternator

12

MREF Program ------

Phase III - Vehicle Demonstrator

• 7 kW of Power

• 9 kW Peak Power

• 30 kW-hrs ofH2 storage

• 28 VDC & 120 V AC formats

• Standard alternator

• Inion a Light Armored Vehicle (Stryker) ~ .. 7 '"

13

s ~ ~ b1J 0 ~

~ 0 s OJ

0 (j)

·c ~ $

fii~ CJ)

~ j ~ --~ . . 0.. ...-...-

r:rJ. ..c

SunLine Demo Program • Contractors

• SunLine Transit Agency • Southwest Research Institute

• Baseline vehicle • Peterbilt Class 8 semi-truck wi Cummins ISL pre-2002 diesel engine

(no EGR) • Fuel cell APU

• Two Hydrogenics PEM units (20 kW total) • On-board hydrogen storage

• Three 5000 psi Dynetek composite units (5 kg total) • Electrified systems to date

• Water'pump, air compressor, engine radiator fan, air conditioning system -

• Other systems that can be electrified include • Pow~r steering, oil pump, inte.g~ated motor starter generator (ISG),

electric turbocompoundirrg _:" 15

FL/Ballard Demo Program

• Baseline vehicle • Freightliner Class 8 semi-truck

• Fuel cell APU • Ballard PEM wi on-board methanol reforming

(2.2 kW total)

• On-board methanol storage

~.

17

Clean Fuels Initiative •

•

• •

•

Today US imports 57 % of oil requirements • Expected to rise to> 70 ~o by 2025

China, India, andE. Europe increase = 54 % • F our times greater than production capability

World consumption from 70 to 121 MM BPD No new CONUS refineries built in 25 yrs • No new us refineries expected to be built (NIMBY) • 88 % refining capacity in 8 states on Great Lakes, East, Gulf, and West

Coasts • Mega-complexes in Texas, Philadelphia, New Jersey, Los Angeles, and

Louisiana o • These are all potential terrorist targets ----.,,;-....

Total supply exhausted in a few decades? • (data from US Energy Information Administration)

18

Clean Fuels Initiative • Use secure indigenous sources of energy

• Coal and petcoke

• Use an environmentally sensitive process to produce a better fuel . • Fischer-Tropsch process

I

• Use coal and petcoke as feed stocks

19

Fischer-Tropsch Fuels • Non-detectable levels of Sulfur, Aromatics, or

Metals

• High cetane number (> 74 vs. 45 - 50) • Very low toxicity / bio-degradeable • Significantly lower emissions (NOx, PM, CO) • Compatible with existing fuel distribution

infrastructure and legacy fleet

• Reduced visible smoke • Great source of Hydrogen

20

JP-8 Reforming Three common methods for reforming -• Partial Oxidation (POX)

• CnHm + ~ n 02 ---+ n CO + 12 m H2 • Low H2 yield, exothermic' reaction, mass transfer limited

• Steam Reforming (SR) • CnHm + n H20 ---+ n CO +- (n + 12 m) H2 • High H2 yield, endothermic, heat transfer limited

• Autothermal Reforming (A TR) • CnHm + X 02 + (2n -- 2x} H 20 ---+ n CO2 + (2n - 2x + 12 m)

H2 • Medium H2 yield, tunable heat duty

.~,

21

JP-8 Reforming '.

• Biggest issue in using JP-8 for fuel cell is sulfur content

• JP-8 surrogate compounds are very difficult to reform due to molecule size • Dodecane and Tetralin are surrogates

• JP-8 reformate better suited for SOFC rather thanPEMFC • > 10 ppm CO not a problem for SOFC

22

JP-8 Refonning· Develop reforming options for SOFC APUs

• Direct catalytic decomposition of JP-8 into C and H2 at 600 -. 800D e • One reactor vs. three (reforming, WGS,

PROX)

• Alternative, novel catalysts needed that will remain active over a wide range of fuels

• Yttria-stabilized zirconia (YSZ),Zirconia-doped ceria (CeZr02), Gadolina-doped ceria (GDC), Gd2 Ti20 7

• Others?'" .... 23

JP-8 Refonning Investigate the feasibility of GTL fuel processes for

APUs

• Develop a "design~r" fuel "from the F -T reaction which meets exact specifications

• Results in an easy-to-reform fuel without pre-g

processing

• Develop a method to selectively remove aromatics as well as sulfur

• Results in an easy-to-reform fuel 24

JP-8 Reforming Develop technical approaches for overcoming barriers

to the success of a JP-8 - fueled SOFe APU

• Use Density Functional Theory calculations to develop novel catalyst / electrode combinations

• Find those material combinations that are resistant to coking and sulfur poisoning as well as having a lower overpotentialloss than Ni

• Determine anode catalyst deactivation mechanisms

• Coke formation is biggest challenge 25

#15383 Sattler 1#15383 Sattler 2